Journal of Heredity 2004:95(4):327–331 ª 2004 The American Genetic Association

doi:10.1093/jhered/esh052

Genetic Diversity and Relationships

in Native Hawaiian Saccharum
officinarum Sugarcane
S. SCHENCK, M. W. CREPEAU, K. K. WU, P. H. MOORE, Q. YU, AND R. MING
From the Hawaii Agriculture Research Center, 99-193 Aiea Heights Dr., Aiea, HI 96701

Downloaded from https://academic.oup.com/jhered/article/95/4/327/2187524 by guest on 06 August 2020

(Schenck, Crepeau, Wu, Yu, Ming); and USDA-ARS, Pacific Basin Agricultural Research Center,
99-193 Aiea Heights Dr., Aiea, HI 96701 (Moore).

Address correspondence to S. Schenck at the above address, or e-mail: sschenck@harc-hspa.com.

Abstract
Commercial sugarcane hybrid cultivars currently in production are high-yielding, disease-resistant, millable canes and are the
result of years of breeding work. In Hawaii, these commercial hybrids are quite distinct from many Saccharum officinarum
canes still in existence that were brought to the islands and cultivated by the native Polynesians. The actual genetic
relationships among the native canes and the extent to which they contributed to the commercial hybrid germplasm has
been the subject of speculation over the years. Genetic analysis of 43 presumed native Hawaiian S. officinarum clones using
228 DNA markers confirmed them to be a group distinct from the modern hybrid cultivars. The resulting dendrogram
tended to confirm that there were several separate S. officinarum introductions that, owing to selections of somatic mutations,
diverged into a number of cluster groups. When the ‘‘Sandwich Isles’’ were discovered by Captain James Cook in 1778, the
Hawaiians were found to be growing sugarcane, S. officinarum (Cook 1785). Sugarcane (ko, in the Hawaiian language)
appeared in a variety of stalk and leaf colors, often with stripes (the ‘‘ribbon canes’’). In the interest of preserving this
historic germplasm, a collection was assembled in the 1920s by Edward L. Caum of the Hawaiian Sugar Planters’
Association and W. W. G. Moir of American Factors. Histories and descriptions of the canes were reported by Moir (1932).

Moir (1932) arranged the native Hawaiian cultivars into
groups and families based purely on their morphological
characteristics (Table 1). Wilfong (1883) stated that Ualalehu,
Ualalehu maoli (native), Honuaula, Laukena (Laukona), Kea
(Kokea), Papa, and Ohua were indigenous natives, while
Lahaina, Palani, Hou, Manulele, Uala, and others were
brought here from abroad. Although Lahaina has a native
Hawaiian name, it was imported from the Marquesas in 1854
(Wilfong 1883; Hawaiian Sugar Planters’ Association re-
cords). Mangelsdorf (1956) surmised that the native canes
might all be selections of somatic mutants of a single Saccharum
officinarum introduction, but this is lost to history. This project
attempted to clarify the actual genetic relationships of the
native Hawaiian S. officinarum varieties that still exist and
their relationship, if any, to the commercial Saccharum hybrids.
Early attempts to profitably produce sugar from the
original Hawaiian canes proved unsuccessful because they
were too soft for milling and were susceptible to introduced
diseases (Mangelsdorf 1956). The native Hawaiian S.
officinarum canes were thought to be infertile, although they
do produce flowers and recent research has shown that
a percentage of the pollen of some of them is viable (Nagai et
al. 1990). Other S. officinarum cultivars and interspecific
hybrids were imported for commercial production and for
breeding (Mangelsdorf 1956). As a result, modern commer-
cial Hawaiian sugarcanes are hybrids of S. officinarum,
Saccharum robustum, Saccharum barberi, and Saccharum spontaneum
species. Detailed records were kept and the parentage of the
present commercial canes can be traced back to the original
imports (Tew 1987). Mangelsdorf (1956) stated that ‘‘All of
the present major varieties of Hawaii include in their ancestry
32-8560, itself until recently the leading variety in the
Territory. The mother of 32-8560 is the Indian variety CO
213, a seedling of P.O.J. 213. The father of 32-8560 is the
Java variety P.O.J. 2878.’’ Other early cultivars included
Lahaina (S. officinarum), H-109 (a progeny of Lahaina and an
unknown parent), and Yellow Caledonia (S. officinarum
imported in 1881 from an unknown location) (Tew 1987).
All of these cultivars still exist and were included in this
study. It is not known whether any of the native Hawaiian S.
officinarum cultivars contributed germplasm to the commer-
cial breeding program.

327
Journal of Heredity 2004:95(4)

Table 1. Groups and families of S. officinarum based on Moir DNA Extraction

(1932)
DNA was extracted from leaf blade tissue that was chopped,
Group I Group II lyophilized, and ground to powder. Genomic DNA
Akilolo family Laukona family extraction was slightly revised from Tai and Tanksley
Nanahu Laukona (1990). Each sample was digested with EcoRI and HindIII.
Pakaweli Lahi About 7.5 lg of DNA per lane were run on a gel. Southern
Akoki family Uahiapele family blotting, radioactive labeling, and autoradiography were as
Akoki Uahiapele described (Chittenden et al. 1994).
Uala Pohina
Manulele family Kea family
Manulele Kea (Kokea)
DNA Probes
Honuaula Miscellaneous
Awela family (seedlings, mutants or imports) Thirty-eight genomic and cDNA probes derived from
Awela (Puaole) Hinahina sugarcane and sorghum were selected for fingerprinting the

Downloaded from https://academic.oup.com/jhered/article/95/4/327/2187524 by guest on 06 August 2020

Uluhui Moano
Miscellaneous Lehu (closely related to above families)
Ohia
Waiohia
The taxonomy and speciation within the genus Saccharum
is complex and uncertain. Although most authors agree that
S. spontaneum and S. robustum are ancestral wild species, and
the whole group—S. spontaneum, S. robustum, S. officinarum,
S. barberi, and Saccharum sinense—is interfertile (Berding and
Roach 1987; Irvine 1999). Chromosome number and ploidy
vary greatly and there is much overlap between them. Recent
advances in genetic technology, notably the application of
restriction fragment length polymorphism (RFLP) markers,
has begun to shed some light on this complex and diverse
genus ( Jannoo et al. 1999; Lu et al. 1994a,b). Interspecific
crosses were the basis for the development of the modern
sugarcane industry, with its many high-yielding, disease-
resistant cultivars adapted to various environmental
conditions worldwide.
Materials and Methods
Sugarcane Cultivars
The sugarcane cultivars used in this project were taken from
plots in the Hawaii Agriculture Research Center (formerly
Hawaiian Sugar Planters’ Association) breeding station
at Maunawili, Hawaii, or from the Waimea Arboretum,
Haleiwa, Hawaii, and other locations around the islands. The
cultivars with Hawaiian names such as Akoki, Pakaweli,
Manulele, etc., are the same clones as those collected by Moir
and Caum (Moir 1932). Their appearance matches Moir’s
description. The cultivars collected from Waimea Arboretum
were previously given to it by Hawaiian Sugar Planters’
Association and are also the same clones that were collected
by Moir and Caum and maintained since that time. Others
were found in various locations and their original names are
unknown. A total of 43 Hawaiian S. officinarum accessions and
three later S. officinarum imports were collected for DNA
fingerprinting, along with two early Saccharum hybrids (circa
1925), five current commercial hybrids, and one sample each
from S. robustum and S. spontaneum added as outgroups.
native Hawaiian sugarcane collection, including 28 sugarcane
cDNA probes derived from cell suspension culture,
germinating lateral buds, and germinating set roots; 1
sugarcane genomic probe; and 9 sorghum genomic probes.
The sorghum genomic DNA probes were obtained from the
Center for Applied Genetic Technology, University of
Georgia, Athens, Georgia. These probes were distributed
throughout the sugarcane genome based on their position on
the Saccharum consensus genetic map (Ming et al. 2002).
Data Analysis
For each probe, the polymorphic restriction fragments
found from all varieties were numbered from high to low
molecular weight. Each fragment (marker) was scored as 1
(present) or 0 (absent) in a spreadsheet. The data were
formatted for the NTSYSpc (version 2.1) cluster analysis
software (Exeter Software Co., Setauket, NY). Mono-
morphic markers were not scored. The RFLP marker data
were used to compute pairwise Dice coefficients (Dice
1945). Cluster analysis was performed on the similarity
matrix using the ‘‘unweighted pair group method using
arithmetic means’’ (UPGMA) algorithm (Sneath and Sokal
1973) provided in the software package NTSYSpc. The
cophenetic correlation coefficient was calculated to test the
goodness-of-fit between the similarity and the cophenetic
matrices. A 50% majority rule consensus tree was calculated
from the most parsimonious trees using PAUP 4.0
(Swofford 2002). Bootstrap values were calculated from
100 replicates.
Results and Discussion
A total of 228 polymorphic DNA markers were detected by
the 38 sugarcane and sorghum DNA probes and scored for
phylogenetic analysis. Each probe detected a discrete
hybridization pattern and produced one to nine polymorphic
markers with an average of six markers per probe. Each
probe detected 5 to 17 DNA fragments, with most of them
being polymorphic. One hundred eighty-five DNA frag-
ments were too close to be discretely scored. Thirty
monomorphic markers were detected by 26 probes. A few
markers showed differences in the band intensity, possibly

328
 Schenck et al.
Genetic Diversity and Relationships in S. officinarum

Table 2. Supposed origins of S. officinarum clones in Hawaii Table 2. Continued

Akilolo (no S. officinarum introduced by Akilolo (no S. officinarum introduced by

longer exists) original Hawaiian immigrantsa longer exists) original Hawaiian immigrantsa
Akoki S. officinarum introduced by Waimea S. officinarum
original Hawaiian immigrantsa,b Waiohia Native Hawaiian S. officinaruma
Cavengerie (Ieie) Native Hawaiian S. officinaruma Yellow Caledonia Introduced in 1881 for breedingc
H52 Early commercial Saccharum
a
hybrid, progeny of White Moir (1932).
Mexicanc b
Kamakea (1872).
H109 Early commercial Saccharum c
HSPA breeding records.
hybrid, progeny of Lahainac d
Wilfong (1883).
Halalii Native Hawaiian S. officinaruma,b
HC62 Unknown Hawaiian S. officinarum
HC63 Unknown Hawaiian S. officinarum reflecting the dosage effect. This type of marker was not

Downloaded from https://academic.oup.com/jhered/article/95/4/327/2187524 by guest on 06 August 2020

HC71 Unknown Hawaiian S. officinarum included in the cluster analysis.
Honaunau-1 S. officinarum Forty-six accessions of S. officinarum clones thought to be
Honaunau-2 S. officinarum Hawaiian natives or early imports were collected and
Honomalino-1 S. officinarum
Honomalino-2 S. officinarum analyzed (Table 2). The numbers following some of the
Honuaula Native Hawaiian S. officinarum, names indicate individual accessions. Clones with the same
mutant of Manulelea,d name should therefore be alike genetically, but this was not
Ieie (Cavengerie) Native Hawaiian S. officinaruma always the case. In some samples there was no known name:
Kalaoa S. officinarum HC (for ‘‘Hawaiian cane’’), unknown, and Maui cane are
Keauhou S. officinarum some of these. In addition, we also included some of the
Kokea (Kea) S. officinarum introduced by early imports and two of the first cultivars (H52 and H109)
original Hawaiian immigrantsa,d bred in Hawaii and grown commercially. The dendrogram of
Koula (Ula) S. officinarum cluster analysis based on Dice similarity coefficients is shown
Lahaina Introduced from the
Marquesas in 1854, parent of
in Figure 1. Seven clusters with two or more closely related
H109c accessions can be distinguished with the outgroup samples S.
Lahi (Ualalehu) Mutant of Laukonaa robustum (MOL 5829) and S. spontaneum (Burma). A cluster
Lauloa Native Hawaiian S. officinaruma was defined as sharing 90% or more identical markers. A
Laukona S. officinarum introduced by second dendrogram was generated using Wagner parsimony,
original Hawaiian immigrantsa,b,d and the bootstrap values for all seven clusters were 100
Lehu Introduced by Europeansa (Figure 2).
Mahaiula S. officinarum A cluster of very closely related clones, all with historic
Manulele S. officinarum introduced by Hawaiian names, can be seen at the top of the diagram
original Hawaiian immigrantsa,b
(cluster I). All of the names within this cluster are cited in one
Moano Introduced by Europeansa or more of the references to be of native Hawaiian origin or
Nanahu Native Hawaiian S. officinarum,
mutant of Akiloloa to be mutants of them (Table 2). Only the names Kalaoa,
Not Kokea (original S. officinarum
Keauhou, and Koula are not mentioned by any of the early
name lost) authors. It is possible that cluster I represents a series of
Ohia Native Hawaiian S. officinaruma somatic mutants from a single original introduction by the
Pakaweli Native Hawaiian S. officinarum, ancient immigrants to the Hawaiian islands, as Mangelsdorf
mutant of Akiloloa (1956) suggested.
Pohina Native Hawaiian S. officinarum, Some of the other canes that were thought to be of native
almost identical to Uahiapelea Hawaiian origin shared less than 80% of the markers with the
Pokapua S. officinarum above-mentioned cluster. These include Manulele, Waiohia,
Puaole (Awela) S. officinarum introduced by Laukona-15, Kokea, and Honuaula. Whether we still have
Europeansd; native the correct canes so named or have other misnamed canes is
Hawaiiana unknown. Our Kokea sample is so different that it may not
Uahiapele Native Hawaiian S. officinarum, even be S. officinarum. Moir (1932) cited an ancient Hawaiian
almost identical to Pohinaa
legend concerning Manulele (‘‘flying bird’’) and considered it
Uala Introduced by Europeansd;
to be a native, as did Kamakea (1872). However, our results
native Hawaiian mutant
of Akokia support Wilfong (1883), who listed it as a later import.
Ualalehu (Lahi) Mutant of Laukonaa Other than the ‘‘core’’ group, there is quite a diversity
Uhu Native Hawaiian S. officinaruma among the Hawaiian clones, indicating that they represent
Uluhui Native Hawaiian mutant many different introductions. Most of Moir’s ‘‘family’’
of Awelaa groupings do appear to be justified genetically (Table 1).
Moir grouped Nanahu and Pakaweli, Akoki and Uala, Uluhui

329
Journal of Heredity 2004:95(4)
Figure 1. Phenogram based on simple matching similarity
coefficients among 43 presumed native Hawaiian S. officinarum
accessions, 3 later S. officinarum imports (Yellow Caledonia,
Lahaina, Kokea), 2 early Hawaiian commercial hybrids (H52,
H109), 5 modern commercial hybrids (H65-7052, H78-7750,
H78-4153, H77-4643, H85-7421), and 2 wild species—S.
robustum and S. spontaneum—as outgroups. The cophenetic
correlation coefficient r ¼ 0.95.
and Puaole, Ohia and Lauloa, and Lahi and Laukona.
However, Uahiapele and Pohina are not clustered together.
Uahiapele is genetically unrelated to Uahiapele-50, nor are
they similar in appearance. It is unlikely that our so-called
Uahiapele is actually the correct clone. Moir included
Manulele and Waiohia among the ancient Hawaiian clones,
whereas our cluster analysis showed that Manulele and
Waiohia shared more than 90% of their markers with the
later import, Yellow Caledonia (cluster V). As expected, the
known foreign imports of S. officinarum, Lahaina and Yellow
Caledonia, as well as the early commercial hybrids H52 and
H109, share less than 80% of their markers with the ‘‘core’’
cluster. Lahaina and H109 share nearly 80% of their markers
and are known to be parent and progeny.
Cluster analyses of the modern commercial Hawaiian
Saccharum hybrids show them to be clearly distinct as a group
from the native Hawaiian canes. Kokea, as well as the early
Hawaiian commercial hybrid H52, also segregates with the
modern commercials. S. officinarum is thought to have
originated from S. robustum, although probably not directly
(Lu et al. 1994b). It is clear from the dendrogram that S.
officinarum is more closely related to the S. robustum clone than
it is to the S. spontaneum clone tested in this study.
330
Downloaded from https://academic.oup.com/jhered/article/95/4/327/2187524 by guest on 06 August 2020
Figure 2. A 50% majority rule consensus tree obtained from
the most parsimonious trees based on amplified fragment
length polymorphism (AFLP) markers. Accessions are identical
to those in Figure 1. The cluster numbers followed the
designation in Figure 1.
One of our hypotheses was that the native Hawaiian
canes represented an isolated and distinct group of S.
officinarum that were nearly identical to each other, but
significantly different from all other S. officinarum clones.
While there was a closely related cluster of Hawaiian clones
(cluster I), most were genetically quite diverse. The results
tended to confirm that there were several separate S.
officinarum introductions into Hawaii that, owing to selections
of somatic mutations, diverged into a number of cluster
groups. Mangelsdorf (1956) stated that the ancient Hawaiian
canes were infertile and therefore could not have contributed
to the germplasm of the commercial cultivars. Our results
support, but do not prove, this conclusion.
Acknowledgments
We thank David Orr for providing sugarcane samples and Andrew Paterson
for providing sorghum genomic DNA probes. This work was supported by
a USDA-ARS agreement (no. CA 58-5320-9-105) with the Hawaii
Agriculture Research Center and support from the Hawaiian Sugar
Technologists.

References
Berding N and Roach BT, 1987. Germplasm collection, maintenance, and
use. In: Sugarcane improvement through breeding (Heinz DJ, ed).
Amsterdam: Elsevier; 143–210.
Chittenden LM, Schertz KF, Lin Y-R, Wing RA, and Paterson AH, 1994. A
detailed RFLP map of Sorghum bicolor 3 S. propinquum, suitable for high-
density mapping, suggests ancestral duplication of Sorghum chromosomes or
chromosomal segments. Theor Appl Genet 87:925–933.
Cook J, 1785. Voyage to the Pacific Ocean, vol. II, book III: 193, 244, 3rd
ed. (vol. VII). London: H. Hugh.
Dice LR, 1945. Measures of the amount of ecologic association between
species. Ecology 26:297–302.
Irvine JE, 1999. Saccharum species as horticultural classes. Theor Appl Genet
98:186–194.
Jannoo N, Grivet L, Seguin M, Paulet F, Domaingue R, Rao PS, Dookun A,
D’Hont A, and Glaszmann JC, 1999. Molecular investigation of the genetic
base of sugarcane cultivars. Theor Appl Genet 99:171–184.
Kamakea DK, 1872. Memoirs of the Bernice Pauahi Bishop Museum.
V(III):582–588.
Lu Y-H, D’Hont A, Paulet F, Grivet L, Arnaud M, and Glaszmann JC,
1994a. Molecular diversity and genome structure in modern sugarcane
varieties. Euphytica 78:217–226.
Lu Y-H, D’Hont A, Walker DIT, Rao PS, Feldmann P, and Glaszmann JC,
1994b. Relationships among ancestral species of sugarcane revealed with
RFLP using single copy maize nuclear probes. Euphytica 78:7–18.
Schenck et al.
Genetic Diversity and Relationships in S. officinarum
Mangelsdorf AJ, 1956. Sugar cane breeding: in retrospect and in prospect.
Proc IX Congr ISSCT; 560–575.
Ming R, Liu SC, Bowers JE, Moore PH, Irvine JE, and Paterson AH, 2002.
Construction of a Saccharum consensus genetic map from two interspecific
crosses. Crop Sci 42:270–283.
Moir WWG, 1932. The native Hawaiian canes. Bulletin 7. Proc IV Congr
ISSCT; 1–8.
Nagai C, Tew T, and Jong J, 1990. Pollen viability and self-incompatibility
of Saccharum officinarum germplasm. Hawaiian Sugar Planters’ Association
Annual Report; 11–12.
Sneath PHA and Sokal RR, 1973. Numerical taxonomy. San Francisco:
W.H. Freeman.
Swofford DL, 2002. PAUP, phylogenetic analysis using parsimony (and
other methods), version 4.0. Sunderland, MA: Sinauer Associates.
Downloaded from https://academic.oup.com/jhered/article/95/4/327/2187524 by guest on 06 August 2020
Tai TH and Tanksley SD, 1990. A rapid and inexpensive method for
isolation of total DNA from dehydrated plant tissue. Plant Mol Biol Rep
8:297–303.
Tew T, 1987. New varieties. In: Sugarcane improvement through breeding
(Heinz DJ, ed). Amsterdam: Elsevier Press; 559–594.
Wilfong GW, 1883. Varieties of cane. Planters Month 2:116–117.
Received May 1, 2003
Accepted April 1, 2004
Corresponding Editor: James Hamrick
331